Access Type

Open Access Dissertation

Date of Award

January 2014

Degree Type


Degree Name



Biochemistry and Molecular Biology

First Advisor

Zhe Yang


Protein X-ray crystallography is a powerful approach for elucidating protein structure and function. The high-resolution data generated by X-ray allow us to visualize protein structures in a three-dimensional (3D) space, which is vital for our understanding of the protein intra- and intermolecular interactions that explain the mechanisms of various biological events. More importantly, such information can provide a structural basis for developing new methods and strategies of targeted drug discovery. In this dissertation, by using X-ray crystallography as the primary approach, we have performed the structural and functional studies of SMYD2 and NHERF1 and have determined their mechanisms of action in epigenetic regulation and protein scaffolding, respectively.

Primarily identified as a histone lysine methyltransferase, SMYD2 has been shown to be play important roles in muscle development and tumorigenesis. In addition to histone substrate, SMYD2 can also methylate non-histone proteins including p53, retinoblastoma tumor suppressor and estrogen receptor alpha. However, there are still many gaps in knowledge regarding the mechanisms underlying the activity regulation and substrate recognition of SMYD2. In this dissertation, we solved the crystal structures of SMYD2 with two different cofactors. Both cofactor-bound SMYD2 structures have a two-lobed structure with the active site partially blocked by a domain at the C-terminus (CTD). Although the two structures are highly superimposable, detailed structural analysis revealed the significantly different CTD conformations, suggesting the CTD flexibility that may be involved in the regulation of SMYD2 histone methyltransferase activity. In addition, the structural similarity between the CTD and the tetratricopeptide repeats (TPR) suggests a possible mechanism for the Hsp90-mediated SMYD activity enhancement. Based on such knowledge, we then employed the co-crystallization approach to study the mechanisms for the substrate recognition. We have successfully co-crystallized SMYD2 with a non-histone substrate, estrogen receptor alpha (ERα). The complex structure revealed that ERα peptide binds SMYD2 in a U-shaped conformation with the binding specificity determined predominantly by residues C-terminal to the target lysine. The structure also showed that the broad specificity of SMYD2 is achieved by multiple molecular mechanisms such as distinct peptide binding modes and the intrinsic dynamics of peptide ligands. Interestingly, a novel potentially SMYD2-specific PEG binding site is identified in the CTD, implicating possible functions in additional substrate binding or protein-protein interactions.

The formation of CXCR2—NHERF1—PLCβmacromolecular complex plays vital roles in both inflammation and pancreatic cancers. In neutrophils, this NHERF1-mediated macromolecular complex is essential in intracellular calcium mobilization and neutrophil migration. In pancreatic cancer cells, this complex regulates tumor proliferation and invasion. Therefore, targeting this NHERF1-mediated macromolecular complex will have great clinical importance. The second objective of this dissertation is to provide the structural basis for the formation of this NHERF1-mediated macromolecular complex. To achieve this, we first solved the complex structures of the NHERF1 PDZ1 domain with the C-terminal sequence of CXCR2 in two different crystal forms. Although the superposition revealed a high degree of overall structural similarity, distinct conformations were observed between the two forms in substrate-binding pocket and bound peptide. These conformational differences indicated that the flexibility of the ligand-binding pocket might be required for diverse peptide recognition. The structural comparison also reveals that the intrinsic dynamics of the peptide ligand may allow the PDZ1 domain for interactions with different peptide recognition residues.

The interactions between NHERF1 and the CXCR2 downstream effector PLCβ3 have been studied using the same strategy as mentioned above. The structural studies of the PDZ1—PLCβ3 complex allowed us to identify the determinants of the PDZ1 binding specificity. We also showed that PLCβ3 can bind PDZ2 in pancreatic cancer cells, consistent with the observation that the peptide binding pocket of these PDZ domains are highly structurally conserved.

In summary, the studies preformed in this dissertation have revealed new insights into the mechanisms behind the lysine methylation machinery and protein scaffolding which are central to many biological processes and diseases. Such findings will be of great benefit in the development of alternative therapeutic strategies and drug design.

Included in

Biochemistry Commons